Review of MRI brain tumor segmentation and MGMT promoter classification methods on BraTs dataset based on Deep learning

• Published in: 2024 IEEE 7th International Conference on Advanced Technologies, Signal and Image Processing (ATSIP) Vol. 1; pp. 249 - 254
• Main Authors: Amor, Feriel, Mzoughi, Hiba, Njeh, Ines, Slima, Mohamed Ben
• Format: Conference Proceeding
• Language: English
• Published: IEEE 11.07.2024
• Subjects: Accuracy; BraTS 2022; BraTS2021; Deep learning; Deep Learning (DL); Gliomas; Image segmentation; Learning systems; Magnetic resonance imaging; MGMT; Performance evaluation; Resonance Imaging (MRI); Reviews; Segmentation
• ISSN: 2687-878X
• DOI: 10.1109/ATSIP62566.2024.10638990

During clinical decision-making, the segmentation process Is considered an important task because it provides valuable information about the location, size, and characterization of the brain tumor. However, manual human segmentation is error-prone, time-consuming, and requires skilled neuroradiologists.Magnetic resonance imaging (MRI) could give extremely detailed images for the investigation and diagnosis of glioblastoma brain cancers.We compared evaluated approaches on the BRATS 2021 and BRATS 2022 datasets and found that they outperformed and could compete with state-of-the-art algorithms in comparable settings.In our research, we focused on two crucial tasks: segmentation and MGMT classification. This study also addresses asn objective evaluation through performance evaluation of cutting-edge DL-based techniques for MR image analysis (Brats 2021-Brats 2022). Based on the findings of the contrasted methods, we can confirm that using a combination of DL techniques will produce more accurate segmentation results than depending on a single, unique methodology. For the second task, five distinct deep learning-based methods were evaluated to predict the methylation state of the MGMT promoter.